Skip to main content
Shaping Europe’s digital future
Projects story | Publication

An open e-Infrastructure to support predictive toxicology and risk assessment

The project OpenRiskNet developed an open e-Infrastructure providing resources and services to a variety of communities requiring risk assessment. This infrastructure offers an important alternative to produce chemicals, cosmetic ingredients or nanomaterials without using animal testing.

Group photo of the people working for the EU project OpenRiskNet


The project OpenRiskNet has been funded by Horizon 2020, the research and innovation programme of the European Union. The project ran from December 2016 until November 2019 and received EUR 2,99 million of EU funding.

As the project recently ended, we have asked Thomas Exner, the coordinator of the project ,to tell us more on the achievements of the project.

What was the first and primary objective of OpenRiskNet? Did the project meet its goal? 

Safety testing of drugs, chemicals, pesticides, nanomaterials, cosmetics and other consumer products was heavily based on animal experiments to identify possible adverse health or environmental effects resulting from their use. The European Union committed to reduce such testing to the absolute minimum following the 3R principles: Replacement, Reduction and Refinement of animal testing. For cosmetics, a complete ban of new animal tests was enforced in 2013 and similar bans are to be expected for other product groups in the near future. This has resulted in a paradigm shift in the area of toxicology and risk assessment, from a phenomenological to a mechanistic discipline relying on in vitro and in silico approaches. The experimental testing is not performed anymore in living species (in vivo) but uses cell culture models (in vitro) or is replaced by computation models (in silico). This transition was supported by the European Commision with specific Horizon 2020 calls. The resulting EU funded projects,e.g. EU-ToxRisk and different NanoSafety projects, have developed so-called new approach methodologies (NAMs) representing important alternatives to classical animal testing applied to the evaluation of chronic and systemic toxicity.

However, no single NAM is predictive for a specific adverse effect and they have to be combined to test batteries and integrated approaches for testing and assessment. This results is an extreme increase in the requirements put on data management, processing, analysis, modelling and simulations to make the output of the different experimental results and computational approaches sharable and reusable.

OpenRiskNet was created to address these needs with the objective to provide an open e-infrastructure that makes existing resources and services available to a variety of communities and domains requiring risk assessment (e.g., chemicals, cosmetic ingredients, drugs or nanomaterials) in an easier-to-use, harmonised and interoperable way. Large databases and highly sophisticated methods, algorithms and tools are available for different tasks such as hazard prediction, toxicokinetics or in vitro-in vivo extrapolations. However, these services are developed independently and provided by different groups world-wide and there is no standardised way to access the data or run modelling workflows. Therefore OpenRiskNet was implemented with the scope to play an important role in developing the concepts and approaches for bringing all these tools together onto a common platform free to use by the providers as well as users of the services.

OpenRiskNet developed an integrated, secure, service-driven and sustainable infrastructure for data management, data sharing, processing, analysis, information mining and modelling. The development and sharing of workflows for analysis, visualisation and reporting were also supported. OpenRiskNet e-infrastructure supports several aspects of risk assessment and safe-by-design product development by allowing integration of toxicology-related data sources, implementation and execution of processing and analysis pipelines and execution of modelling workflows for pharmacokinetics or pharmacodynamics. OpenRiskNet activities improved the accessibility, interoperability and standardisation of existing services, interconnected them and  developed protocols for their deployment on all relevant hardware as virtual research environments including aspects of data security. These activities were aligned to the FAIR data principles and guidelines in order to make data Findable, Accessible, Interoperable and Reusable.

Finally, case studies implemented and demonstrated during the project offered concrete examples on how the services available in the e-infrastructure can be used. These case studies have been developed to allow automation and functional integration for applications in hazard identification and risk assessment. Each of them addressed specific areas of the risk assessment process  where automation is highly desirable, like prediction of metabolism, model development for hazard prediction, analysis of transcriptomics data and mapping data to adverse outcome pathways. The resulting modules are fine-tuned for the utilisation and application of new approach methodologies in order to accelerate the replacement of animals in risk assessment scenarios and can now be applied and combined for the full assessment of specific molecules and nanomaterials by academic and industry researchers and risk assessors e.g., as part of the research projects mentioned above.

How the outcomes of OpenRiskNet will be harnessed now that the project ended?

OpenRiskNet was working together with its stakeholders, related projects and communities, as well as with pan-European research activities and e-infrastructures to fully collect and address the users requirements and needs, foster adoption, as well as align with other major research, innovation and infrastructure activities for optimising mutual benefits.

OpenRiskNet data resources and software services were aligned with the activities of international projects leading the development of open standards for predictive toxicology resources like the EU NanoSafety Cluster, EU-ToxRisk, PhenoMeNal, NanoCommons and the US Nano Working Group and at a larger scale with ELIXIR and the European Open Science Cloud (EOSC), including here OpenAIRE and eInfraCentral. The exchange with these groups ensures that OpenRiskNet reused standards wherever possible and that OpenRiskNet also contributes to standards by putting forward project results to the most suitable international bodies.

Moreover, OpenRiskNet services were integrated in the EOSC catalogue of services, activity that supports now the effort in increasing the visibility of the risk assessment group/services within the larger EU e-infrastructure community. This established also the premises for future joint activities and developments aligned with the international initiatives in this area. The successful evaluation for technical feasibility and impact of OpenRiskNet within the EOSC Early Adopter Programme is one example in this direction.

The main aim is now to promote and sustain OpenRiskNet as a major European provider of e-services for the toxicology and neighboring communities, and also as a place to offer and purchase data, predictive toxicology and risk assessment services following a marketplace concept and aligned to or even as part of the EOSC marketplace.

Is EU funding important for the European research? How has it contributed to OpenRiskNet success?

The paradigm shift in toxicology imposed by new regulations resulting from increased public awareness and pressure for animal welfare is only possible as a collective effort of all stakeholders from academia, industry and regulatory agencies. There are quite some challenges to be overcome before regulatory acceptance of NAMs for all important toxicology endpoints will be achieved, best addressed by public-private partnerships supported by the EU through Horizon 2020 and Horizon Europe, the next research and innovation framework programme as well as the Innovative Medicines Initiative (IMI) and other national and European funding possibilities. However, sharing and reuse of approaches between individual projects and knowledge transfer is needed to optimise the impact and guarantee sustainability best organised by independent (e-)infrastructure projects. The EU funding available to the OpenRiskNet consortium during the 3-year project supported the interoperable implementation of approaches from the risk assessment and related areas, partially sustaining developments from previous initiatives like OpenTox, SEURAT-1 and eNanoMapper coupled with new innovative solutions designed and developed during the project. Thus, the existing community standards for data and services could be re-used and enhanced, all supporting risk assessment and safe-by-design studies. Besides the important contribution to the technical and scientific developments for the field covered by OpenRiskNet, the funding scheme allowed us to outreach to additional communities, engage directly with stakeholders and integrate additional services not completely covered by the consortium. This was facilitated by the Associated Partners and Implementation Challenge programmes, instruments used to select and award additional partners and third-party services joining the initiative and enriching the infrastructure.

The project successfully managed to raise awareness of many state-of-the-art processing, analysis and modelling services by enhancing their accessibility, and contributed towards establishing best-practices on how to share data and achieve interoperability of online tools. And finally, the project demonstrates the features and functionality of the e-infrastructure using real-world applications to foster its worldwide adoption, large resources were invested in a range of outreach, dissemination, networking and training activities.